Epidemiological Survey of Human Metapneumovirus Infection in a Large Pediatric Tertiary Care Center
http://www.100md.com
微生物临床杂志 2005年第11期
Laboratoire de Sante Publique du Quebec/Institut National de Sante Publique du Quebec, Sainte-Anne-de-Bellevue, Quebec, Canada
Hpital Sainte-Justine, Montreal, Quebec, Canada
ABSTRACT
The human metapneumovirus (hMPV) was recently identified and linked to acute respiratory tract infections (ARTI). To assess the clinical importance of this virus in infants and children, we developed a rapid and efficient reverse transcription-PCR-based screening method for a large volume of samples and tested retrospectively a collection of 1,132 respiratory specimens submitted over a full year period to the virology laboratory of a large tertiary care pediatric center in Montreal, Canada. A total of 41 samples from 37 patients were positive by this method. During the winter months of 2001, up to 8% of specimens submitted for respiratory virus testing were hMPV positive. Sequencing data of the hMPV M gene revealed that two genogroups of the virus, each of which can be divided into two subgroups, cocirculated during this time period. A case-controlled study was conducted to compare the symptoms associated with hMPV infection with those involving other etiologic agents causing ARTI. Symptoms most frequently observed in hMPV-positive patients were cough, wheezing, and dyspnea, although the symptomatology could differ substantially from patient to patient. No distinct symptom profile could be associated with hMPV. Three nosocomial cases of hMPV infection were identified. Together, our data suggest that hMPV is a significant cause of symptomatic respiratory tract infections in infants and children. The incidence of the disease and the morbidity associated with the infection justify adding hMPV to the list of common respiratory viruses routinely screened for by clinical laboratories.
INTRODUCTION
Infection with respiratory viruses is a common cause of morbidity and mortality in childhood. Despite the use of more-sensitive diagnostic tools by clinical laboratories, an important proportion of respiratory infections still cannot be associated to any known pathogen.
Since the first report of acute respiratory tract infections (ARTI) caused by the human metapneumovirus (hMPV) in 2001 (23), the virus has been detected in a number of countries, suggesting a world-wide distribution (6, 8-10, 13-15, 18-21, 25). hMPV is an enveloped, negative-stranded RNA virus possessing a nonsegmented genome, which has been tentatively classified within the Paramyxoviridae family. The two hMPV genogroups (A and B) that have been characterized to date share an approximate overall identity of 85% at the nucleotide level (20, 23, 24).
In 2002, Peret et al. (20) documented hMPV infection cases in the Canadian province of Quebec and linked the virus to potentially severe illnesses in children. The finding prompted us to investigate the prevalence and the clinical importance of the newly discovered pathogen as a cause of ARTI in the pediatric population. We developed rapid and sensitive reverse transcription-PCR (RT-PCR) assays to detect the virus directly in samples submitted for respiratory virus testing to the virology laboratory of Sainte-Justine Hospital (SJH), a reference pediatric tertiary care center in the city of Montreal, Canada. A total of 41 samples were found positive for the presence of hMPV. Results of a parallel clinical control study suggest that hMPV is an identifiable cause of ARTI in the pediatric population.
MATERIALS AND METHODS
Specimens. The sample series analyzed consisted of 1,132 respiratory specimens submitted throughout the year 2001 to the virology laboratory at SJH (Table 1); an additional 42 samples received during this period were not included in the study because of insufficient residual volume for further testing. A large proportion of samples (>70%) were collected at SJH from patients either hospitalized or seen at the emergency room or at an outpatient clinic as part of the investigation of their illnesses. The remaining specimens were submitted by other public or private laboratories in the province of Quebec in Canada. The series analyzed included swabs (58.6%), nasopharyngeal aspirates (NPA) (31.9%), bronchoalveolar lavage fluid (BAL) specimens (6.5%), and other respiratory specimens (3.0%) such as endotracheal aspirates. Pleural fluid and biopsy specimens were excluded.
Specimens were submitted in either viral transport medium or saline solution for culture, direct immunofluorescence assay, or enzyme immunoassay. Respiratory viruses routinely identified were influenza virus A and B; parainfluenza virus 1, 2, and 3; human respiratory syncytial virus (hRSV); and adenoviruses. Remaining volumes used for this study had been stored at less than –70°C. RT-PCR assays for hMPV were performed in a blind fashion with regards to the patient's medical information.
Case-controlled study. A control group was randomly selected from patients whose samples had tested negative for hMPV and combined with positive patient identifiers on a list which was submitted to two clinical investigators. Seven hMPV-positive patients and four with hMPV-negative results were excluded from the study due to a lack of clinical information. The respiratory samples collected from the control group were retested individually by RT-PCR and confirmed to be hMPV negative. Medical records were examined in a blind fashion, according to a preestablished scheme. The data collection was limited to the information recorded by the treating physicians on medical charts.
The investigators sought the following demographic parameters: age, sex, and postal code. The epidemiologic data included the date of onset of disease, infectious contacts, recent travel history, day care center or school attendance, and hospitalization the week prior to the illness. Underlying medical conditions and medications were reviewed. Investigators also looked for disease manifestations such as fever, rhinorrhea, wheezing, cough, and other signs and symptoms possibly related to the illness.
Results of the following tests were noted: complete blood count, sedimentation rate, chest X-ray, respiratory viral cultures or antigenic detection, nucleic acid detection for Mycoplasma pneumoniae and Bordetella pertussis, and serological studies for specific infectious agents. The need for admission to the intensive care unit, mechanical ventilation, oxygen requirements, and other therapeutic regimens were also recorded.
Nucleic acid extractions. In the first round of screening, all samples except BAL specimens were analyzed in pools of eight samples, prepared by mixing equal volumes of each specimen in the order they were originally submitted to the laboratory. In preliminary analyses, we had found that the efficiencies of recovering and detecting hMPV from respiratory specimens of various natures spiked with in vitro-cultured viruses were similar for all sample types, with the notable exception of BAL fluid specimens (data not shown). Thus, to avoid possible PCR inhibition caused by BAL samples in pools, all BAL specimens were analyzed individually.
The QIAGEN UltraSens kit (QIAGEN, Mississauga, Ontario, Canada), which allows the extraction of total nucleic acids from up to 1 ml of sample volume, was found to be more efficient than several other commercially available nucleic acid extraction kits for recovering hMPV in spiked respiratory specimens (data not shown). Therefore, this method was selected for extracting nucleic acids from pooled samples. In our hands, the QIAamp UltraSens kit coupled with QIAvac purification module 6S gave a higher nucleic acid yield recovery than the centrifugation-based protocol recommended by the manufacturer. Nucleic acids from individual clinical specimens from positive pools were extracted with the QIAamp viral RNA mini kit coupled with QIAvac module 6S as instructed by the manufacturer.
For BAL samples, total RNA was extracted from 200-μl aliquots using the UltraSpec-3 reagent (Biotecx Laboratories, Inc., TX), according to the manufacturer's instructions.
RT-PCR and DNA sequencing. Initial oligonucleotide primer sets were selected on the basis of sequences published by van den Hoogen et al. (23) and available in GenBank. Large genome segments comprising most of the M and F genes could then be amplified and sequenced from eight distinct hMPV isolates cultured in vitro; isolates characterized correspond to CAN98 and CAN99 groups described by Peret et al. in 2002 (20). Subsequent oligonucleotide primers were designed to detect both genogroups.
RT-PCR was performed in a single step using the One-Step RT-PCR kit from QIAGEN as described by the manufacturer. For detecting hMPV in human specimens, one primer pair targeting the M gene and one targeting the M-F intergenic region were used in parallel: Mf1 (5'-ATGTCTGTACTTCCYAA-3') and MReg2 (5'-GCTTATTGCAGCTTCAACAG-3'); and HC 2f (5'-GCACTGTGTGACTTYATGGA-3') and VIRr (5'-ACTTTCCAAGACATTTTTA-3'). Although the first primer pair was found to be more sensitive with serial dilutions of known positive samples, both sets were equally efficient in detecting hMPV in clinical specimens. Cycling conditions were as follows: 50°C for 30 min; 95°C for 15 min; 40 cycles, with 1 cycle consisting of 94°C for 15 s, 52°C for 60 s, and 72°C for 90 s; followed by a final extension at 72°C for 7 min. Amplicons were detected by agarose gel electrophoresis and staining with ethidium bromide.
The hMPV-positive isolates were further characterized by RT-PCR and DNA sequencing using primers targeting the F and M genes. A 1.9-kb fragment spanning the M-F gene region was generated using Mf1 and Fr1 primers (5'-CCYTCAACTTTGCTTAG-3'). Additional primers used for DNA sequencing were as follows: MFIRf (5'-CAARTAAAAATGTCNTGGAA-3'), MF3 (5'-AGAACAGGTTGGTAYAC-3'), HC 7f (5'-AGTAAGAGAGCT GAAAGARTT-3'), and Fr2 (5'-TGCCGCACAACATTTAG-3').
For DNA sequencing, PCR fragments were purified with QIAquick PCR purification columns from QIAGEN as described by the manufacturer, and both strands were sequenced using the BigDye Terminator Cycle Sequencing V 2.0 Ready Reaction kit (PE Biosystems, Foster City, CA). Labeled DNA fragments were separated and detected using ABI genetic analyzer 3100 (Applied Biosystems).
DNA sequence analyses. Base calling and assemblies were performed using the Phed/Phrap software package (University of Washington). Edited sequences were aligned with Clustal W (EMBL, Heidelberg, Germany). Phylogenetic analyses were carried out by using the Phylogeny Inference Package (PHYLIP, version 3.5c; J. Felsenstein, University of Washington) (12). The distance matrix was generated using the DNAdist program (maximum likelihood), and the tree was generated using the neighbor-joining program. Phylogenetic trees were visualized using Treeview (version 1.6.6).
RESULTS
Detection of hMPV in clinical specimens. In 2001, the number of samples submitted for respiratory virus testing to the SJH virology laboratory displayed the usual seasonal pattern. During the winter months, the majority of samples tested positive either by enzyme immunoassay, direct immunofluorescence, and/or tissue culture for hRSV (Fig. 1). Influenza infection peaked in February. On the other hand, adenovirus- and parainfluenza virus-positive samples were found throughout the year.
In our initial screening for hMPV by RT-PCR, the virus was detected in 32 out of 133 pools (24%), and in a single BAL sample out of 74 tested (1%). Specimens from positive pools were then tested separately. Altogether, 41 specimens collected from 37 patients (19 males and 18 females) were found to be positive for hMPV. At least one hMPV-positive sample was found in each positive pool; two and three positive specimens were found in 12.5% and 6.2% of the positive pools, respectively. The two primer sets used for the RT-PCR screening amplified their respective targets from all positive samples.
Similar to hRSV and influenza, the hMPV respiratory infection appeared seasonal, with 37 of 41 positive specimens (90%) collected from January to March (Fig. 2). During this period, 8% of respiratory specimens tested were positive for hMPV.
Sequencing and alignment of specimens. To determine which hMPV genogroup circulated in Montreal, Canada, in 2001, partial sequencing of the M genes of all positive specimens was carried out. Most specimens (73.2%; n = 30) displayed a higher degree of identity with the genogroup B strain Can98-75 (96.3 to 99.6%) than with the prototype hMPV00-1 strain of genogroup A (83.2 to 87.3%) at the nucleotide level (Fig. 3A). Our sequencing data also suggest that each genogroup may be divided into two subgroups. However, no amino acid differences were observed between the two subgroups.
To investigate further the genetic makeup of the circulating genogroups, sequencing of the F gene was undertaken with samples selected from each subgroup. Three samples more closely related to the prototype hMPV00-1 and five samples related to Can98-75 were sequenced. The nucleotide sequence divergence of the F genes of isolates and reference strains was higher than that of the M gene, but the same subgrouping was observed (Fig. 3B).
Clinical manifestations. Clinical data of patients whose respiratory samples tested positive for hMPV were compared to those of patients who tested negative in a case-controlled study. No differences were noted between the two groups with regard to gender, age group, month of the year at disease occurrence, and home address. However, there was a tendency for more chronic illnesses and infectious contacts in group A (hMPV positive) and for more cases of immunosuppression in group B (hMPV negative) (Table 2).
More patients in group A presented with cough and radiographic images of hyperexpansion, peribronchitis, and atelectasis than those in group B. Serological analyses looking for other infectious agents were requested more often in group B than in group A. The duration of admission for hospitalized children did not differ, nor did the duration of stay in the intensive care unit. The need for oxygen and mechanical ventilation or for other treatment was similar in both groups. The final diagnosis was more frequently worded as bronchiolitis or bronchospasm in group A.
Among the 30 hMPV-positive patients, 6 (20%) were coinfected with another commonly identified respiratory virus, but none were coinfected with influenza viruses (Table 3). On the other hand, in the hMPV-negative control group, 13 out of 31 patients (42%) were infected with respiratory viruses likely to be responsible for their symptoms, including 7 who were shown to be infected with an influenza virus. These results are consistent with hMPV being responsible for a certain proportion of the symptoms observed in group A.
In three separate cases of hMPV infection, the exposure was clearly nosocomial. The first case was a 12-year-old girl with encephalopathy and epilepsy who lived in a rehabilitation center. She developed a phlegmy cough, vomiting, and dyspnea with bronchospasm and was admitted to SJH for pneumonia. The second patient was an 11-year-old male with chronic encephalopathy admitted to SJH in January 2000. Approximately 1 year later, while still hospitalized, he developed a viral illness with fever, respiratory symptoms, hyperventilation, and diarrhea. A diagnosis of upper respiratory tract infection was made. The third patient was a 20-month-old infant suffering from congenital lactic acidosis and severe epilepsy. He had been hospitalized since May 2000; in January 2001, he had a fever. The fever resolved, and no precise diagnosis was established at that time. Respiratory samples from the three patients were identified as positive for hMPV by RT-PCR.
Additionally, hMPV contributed to the death of a 3-year-old child handicapped by thoracic asphyxiating dystrophy (Jeune syndrome) who was receiving steroid and oxygen treatment. The patient had been admitted in January 2001 and died 5 days later from multiorganic and respiratory failure. In a separate case, a 9-month-old infant who had traveled to Mali was taken to the emergency room the day he returned to Canada. Symptoms included fever and respiratory and gastrointestinal symptoms of 24-h duration; he had been coughing for 1 week prior to his admission to SJH. This patient had likely been infected with hMPV abroad, while in Africa or on his way back to Canada.
DISCUSSION
We retrospectively assessed the clinical importance of hMPV infection as a cause of ARTI in a pediatric population in Montreal in the Canadian province of Quebec in 2001. We evaluated the incidence over a full year period by RT-PCR testing on nearly all respiratory samples submitted to the virology laboratory of a major pediatric tertiary care center. Samples for which another respiratory virus had already been found were also included. Thus, our approach allowed us to uncover an unusual seasonal pattern and offered the possibility of identifying coinfections. Clinical symptoms associated with hMPV were then compared with those of other respiratory infections in a case-controlled study.
To detect hMPV, we designed RT-PCR assays to amplify viruses of both genogroups and assess the relative sensitivity and specificity of this test using hMPV spiked respiratory samples of different types. Optimized assays were then applied for screening respiratory samples. The rationale for using a molecular approach for this study was based on the following considerations: initial reports of hMPV replication in cell culture suggested that it is a fastidious virus (23); samples analyzed in this study had been subjected to one or more freeze-thaw cycles, for which the effect on hMPV viability was unknown; no antigenic detection was available, and a molecular technique was needed in any case to formally identify the virus. The present study was not aimed at comparing the sensitivity and specificity of our RT-PCR-based assays with those developed by other groups. Nonetheless, it is worth mentioning that the two primer pairs designed for this study detected variants of the two genogroups and that PCR product sequencing confirmed the identity of virus in all hMPV-positive samples.
hMPV was detected in 41 respiratory specimens (3.6% of samples tested) collected from 37 patients. In cases when more than one clinical specimen from a patient tested positive for hMPV, the sequences derived from the specimens were identical. Phylogenic analyses demonstrated a clear dominance of genogroup B in this sample series. As for most respiratory infections, hMPV was mostly found during winter months (90%, or 37 specimens). Interestingly, the virus was not detected in samples collected in November and December, when hRSV was already active (Fig. 1 and 2). In 2001, the same seasonal hMPV occurrence pattern was observed in the neighboring states of the United States and in other regions of Canada (1, 9). In the present study, hMPV was also detected in samples collected in the spring, summer, and autumn, albeit at a low frequency, indicating that hMPV may circulate throughout the year and cause sporadic infections. This finding carries important implications for laboratory investigation of ARTI cases occurring outside the typical respiratory virus season.
NPA was found to be better than throat swabs for hMPV detection. Many respiratory viruses infect the ciliated epithelium of the posterior nasopharynx, and it is generally accepted that NPA collect a superior number of cells than NP washes, NP swabs, and throat swabs do.
The proportion of BAL samples found to be positive for hMPV appeared low at first sight. However, only 21 BAL specimens (28%) were collected between January and March 2001, when hMPV infections peaked, whereas 38% of other specimen types were collected during this 3-month period. Also, we noted a lower efficiency of our hMPV RT-PCR with BAL samples compared with other sample types when spiked with cultured virus (results not shown), a difference that we attributed either to interference with the high content of human nucleic acids in BAL fluids or to residual glutaraldehyde in endoscope channels and in rinsing water that could have impinged on Taq polymerase efficacy (11). Those technical problems may have contributed to an underestimation of the number of BAL fluid samples found positive for hMPV.
Cough (87%), wheezing (40%), and dyspnea (57%) were the symptoms often associated with hMPV infection. Overall, in agreement with the results of other retrospective or prospective studies comparing symptoms of respiratory infections (3, 4, 7, 9, 10, 15-17, 19, 25, 27), our investigation did not reveal a specific symptomatology which could clinically differentiate hMPV from influenza or hRSV infections. Several investigators have found a distinct association between hMPV and bronchiolitis in young children. Our study did not find such a close association, although the terms bronchiolitis and bronchospasm were used more often in the final diagnosis of hMPV-positive cases (40%) than hMPV-negative cases (16%). Moreover, the detection methods used to identify common respiratory viruses are assumed to be less sensitive than the amplification method applied for the detection of hMPV, and we did not seek the presence of other viruses, such as rhinoviruses, coronaviruses, and parainfluenza virus 4, which were likely involved in a certain number of cases. A recent study conducted in France on children admitted for bronchiolitis found RSV to be the predominant causal agent, followed by rhinoviruses and finally by hMPV (5).
Coinfections with hRSV and human parainfluenza viruses were detected (Fig. 3). Mixed viral infections are thought to occur in 3 to 12% of ARTI (26). Thus, our results were not surprising in the context of the peak of respiratory infections during the winter months in Montreal, Canada. In coinfections, a direct contribution of hMPV to an exacerbation of symptoms was not observed, though the low number of cases found during this study preclude any significant conclusion. It is noteworthy that, despite the fact that both groups of patients were recruited from the same period of the year, only patients from the hMPV-negative control group had influenza virus recovered from their respiratory secretions.
This report demonstrates that pooling respiratory samples for hMPV detection is a simple and efficient way for screening a large volume of specimens. This method can be applied for surveillance. This is of particular importance, since the present study and others (9) have documented hMPV infections acquired nosocomially. Therefore, isolation procedures should be taken when hMPV infection is suspected, especially in pediatric units with severely compromised patients and in intensive care units.
Large prospective studies over longer periods of time will eventually help to determine the full clinical spectrum of manifestations associated with this virus, as well as its role in community-acquired and nosocomially acquired respiratory illnesses. Continued surveillance of hMPV infection is warranted to fully understand and to appreciate the clinical importance of this virus.
ACKNOWLEDGMENTS
We thank Andree Falardeau, Jasmine Chamberland, Lyne Desautels, and Regis Cantin for excellent technical assistance; Pierrette Roy for secretarial support; and Donald Murphy, Rejean Dion, Micheline Fauvel, and Peter Barriga for helpful comments and valuable discussions.
REFERENCES
Bastien, N., D. Ward, P. Van Caeseele, K. Brandt, S. H. Lee, G. McNabb, B. Klisko, E. Chan, and Y. Li. 2003. Human metapneumovirus infection in the Canadian population. J. Clin. Microbiol. 41:4642-4646.
Biacchesi, S., M. H. Skiadopoulos, G. Boivin, C. T. Hanson, B. R. Murphy, P. L. Collins, and U. J. Buchholz. 2003. Genetic diversity between human metapneumovirus subgroups. Virology 315:1-9.
Boivin, G., G. De Serres, S. Cote, R. Gilca, Y. Abed, L. Rochette, M. G. Bergeron, and P. Dery. 2003. Human metapneumovirus infections in hospitalized children. Emerg. Infect. Dis. 9:634-640.
Bosis, S., S. Esposito, H. G. Niesters, P. Crovari, A. D. Osterhaus, and N. Principi. 2005. Impact of human metapneumovirus in childhood: comparison with respiratory syncytial virus and influenza viruses. J. Med. Virol. 75:101-104.
Bouscambert-Duchamp, M., B. Lina, A. Trompette, H. Moret, J. Motte, and L. Andreoletti. 2005. Detection of human metapneumovirus RNA sequences in nasopharyngeal aspirates of young French children with acute bronchiolitis by real-time reverse transcriptase PCR and phylogenetic analysis. J. Clin. Microbiol. 43:1411-1414.
Dollner, H., K. Risnes, A. Radtke, and S. A. Nordbo. 2004. Outbreak of human metapneumovirus infection in Norwegian children. Pediatr. Infect. Dis. J. 23:436-440.
Ebihara, T., R. Endo, N. Ishiguro, T. Nakayama, H. Sawada, and H. Kikuta. 2004. Early reinfection with human metapneumovirus in an infant. J. Clin. Microbiol. 42:5944-5946.
Ebihara, T., R. Endo, H. Kikuta, N. Ishiguro, H. Ishiko, M. Hara, Y. Takahashi, and K. Kobayashi. 2004. Human metapneumovirus infection in Japanese children. J. Clin. Microbiol. 42:126-132.
Esper, F., D. Boucher, C. Weibel, R. A. Martinello, and J. S. Kahn. 2003. Human metapneumovirus infection in the United States: clinical manifestations associated with a newly emerging respiratory infection in children. Pediatrics 111:1407-1410.
Falsey, A. R., D. Erdman, L. J. Anderson, and E. E. Walsh. 2003. Human metapneumovirus infections in young and elderly adults. J. Infect. Dis. 187:785-790.
Farina, A., M. H. Fievet, F. Plassart, M. C. Menet, and A. Thuillier. 1999. Residual glutaraldehyde levels in fiberoptic endoscopes: measurement and implications for patient toxicity. J. Hosp. Infect. 43:293-297.
Felsenstein, J. 1997. An alternating least squares approach to inferring phylogenies from pairwise distances. Syst. Biol. 46:101-111.
Freymouth, F., A. Vabret, L. Legrand, N. Eterradossi, F. Lafay-Delaire, J. Brouard, and B. Guillois. 2003. Presence of the new human metapneumovirus in French children with bronchiolitis. Pediatr. Infect. Dis. J. 22:92-94.
Galiano, M., C. Videla, S. S. Puch, A. Martinez, M. Echavarria, and G. Carballal. 2004. Evidence of human metapneumovirus in children in Argentina. J. Med. Virol. 72:299-303.
IJpma, F. F., D. Beekhuis, M. F. Cotton, C. H. Pieper, J. L. Kimpen, B. G. van den Hoogen, G. J. van Doornum, and D. M. Osterhaus. 2004. Human metapneumovirus infection in hospital referred South African children. J. Med. Virol. 73:486-493.
Konig, B., W. Konig, R. Arnold, H. Werchau, G. Ihorst, and J. Forster. 2004. Prospective study of human metapneumovirus infection in children less than 3 years of age. J. Clin. Microbiol. 42:4632-4635.
Mullins, J. A., D. D. Erdman, G. A. Weinberg, K. Edwards, C. B. Hall, F. J. Walker, M. Iwane, and L. J. Anderson. 2004. Human metapneumovirus infection among children hospitalized with acute respiratory illness. Emerg. Infect. Dis. 10:700-705.
Nissen, M. D., D. J. Siebert, I. M. Mackay, T. P. Sloots, and S. J. Withers. 2002. Evidence of human metapneumovirus in Australian children. Med. J. Aust. 176:188.
Peiris, J. S., W. H. Tang, K. H. Chan, P. L. Khong, Y. Guan, Y. L. Lau, and S. S. Chiu. 2003. Children with respiratory disease associated with metapneumovirus in Hong Kong. Emerg. Infect. Dis. 9:628-633.
Peret, T. C., G. Boivin, Y. Li, M. Couillard, C. Humphrey, A. D. Osterhaus, D. D. Erdman, and L. J. Anderson. 2002. Characterization of human metapneumoviruses isolated from patients in North America. J. Infect. Dis. 185:1660-1663.
Stockton, J., I. Stephenson, D. Fleming, and M. Zambon. 2002. Human metapneumovirus as a cause of community-acquired respiratory illness. Emerg. Infect. Dis. 8:897-901.
van den Hoogen, B. G., T. M. Bestebroer, A. D. Osterhaus, and R. A. Fouchier. 2002. Analysis of the genomic sequence of a human metapneumovirus. Virology 295:119-132.
van den Hoogen, B. G., J. C. de Jong, J. Groen, T. Kuiken, R. de Groot, R. A. Fouchier, and A. D. Osterhaus. 2001. A newly discovered human pneumovirus isolated from young children with respiratory tract disease. Nat. Med. 7:719-724.
van den Hoogen, B. G., D. M. Osterhaus, and R. A. Fouchier. 2004. Clinical impact and diagnosis of human metapneumovirus infection. Pediatr. Infect. Dis. J. 23:S25-S32.
Viazov, S., F. Ratjen, R. Scheidhauer, M. Fiedler, and M. Roggendorf. 2003. High prevalence of human metapneumovirus infection in young children and genetic heterogeneity of the viral isolates. J. Clin. Microbiol. 41:3043-3045.
Waner, J. L. 1994. Mixed viral infections: detection and management. Clin. Microbiol. Rev. 7:143-151.
Williams, J. V., P. A. Harris, S. J. Tollefson, L. L. Halburnt-Rush, J. M. Pingsterhaus, K. M. Edwards, P. F. Wright, and J. E. Crowe, Jr. 2004. Human metapneumovirus and lower respiratory tract disease in otherwise healthy infants and children. N. Engl. J. Med. 350:443-450.(Frederic Chano, Celine Ro)
Hpital Sainte-Justine, Montreal, Quebec, Canada
ABSTRACT
The human metapneumovirus (hMPV) was recently identified and linked to acute respiratory tract infections (ARTI). To assess the clinical importance of this virus in infants and children, we developed a rapid and efficient reverse transcription-PCR-based screening method for a large volume of samples and tested retrospectively a collection of 1,132 respiratory specimens submitted over a full year period to the virology laboratory of a large tertiary care pediatric center in Montreal, Canada. A total of 41 samples from 37 patients were positive by this method. During the winter months of 2001, up to 8% of specimens submitted for respiratory virus testing were hMPV positive. Sequencing data of the hMPV M gene revealed that two genogroups of the virus, each of which can be divided into two subgroups, cocirculated during this time period. A case-controlled study was conducted to compare the symptoms associated with hMPV infection with those involving other etiologic agents causing ARTI. Symptoms most frequently observed in hMPV-positive patients were cough, wheezing, and dyspnea, although the symptomatology could differ substantially from patient to patient. No distinct symptom profile could be associated with hMPV. Three nosocomial cases of hMPV infection were identified. Together, our data suggest that hMPV is a significant cause of symptomatic respiratory tract infections in infants and children. The incidence of the disease and the morbidity associated with the infection justify adding hMPV to the list of common respiratory viruses routinely screened for by clinical laboratories.
INTRODUCTION
Infection with respiratory viruses is a common cause of morbidity and mortality in childhood. Despite the use of more-sensitive diagnostic tools by clinical laboratories, an important proportion of respiratory infections still cannot be associated to any known pathogen.
Since the first report of acute respiratory tract infections (ARTI) caused by the human metapneumovirus (hMPV) in 2001 (23), the virus has been detected in a number of countries, suggesting a world-wide distribution (6, 8-10, 13-15, 18-21, 25). hMPV is an enveloped, negative-stranded RNA virus possessing a nonsegmented genome, which has been tentatively classified within the Paramyxoviridae family. The two hMPV genogroups (A and B) that have been characterized to date share an approximate overall identity of 85% at the nucleotide level (20, 23, 24).
In 2002, Peret et al. (20) documented hMPV infection cases in the Canadian province of Quebec and linked the virus to potentially severe illnesses in children. The finding prompted us to investigate the prevalence and the clinical importance of the newly discovered pathogen as a cause of ARTI in the pediatric population. We developed rapid and sensitive reverse transcription-PCR (RT-PCR) assays to detect the virus directly in samples submitted for respiratory virus testing to the virology laboratory of Sainte-Justine Hospital (SJH), a reference pediatric tertiary care center in the city of Montreal, Canada. A total of 41 samples were found positive for the presence of hMPV. Results of a parallel clinical control study suggest that hMPV is an identifiable cause of ARTI in the pediatric population.
MATERIALS AND METHODS
Specimens. The sample series analyzed consisted of 1,132 respiratory specimens submitted throughout the year 2001 to the virology laboratory at SJH (Table 1); an additional 42 samples received during this period were not included in the study because of insufficient residual volume for further testing. A large proportion of samples (>70%) were collected at SJH from patients either hospitalized or seen at the emergency room or at an outpatient clinic as part of the investigation of their illnesses. The remaining specimens were submitted by other public or private laboratories in the province of Quebec in Canada. The series analyzed included swabs (58.6%), nasopharyngeal aspirates (NPA) (31.9%), bronchoalveolar lavage fluid (BAL) specimens (6.5%), and other respiratory specimens (3.0%) such as endotracheal aspirates. Pleural fluid and biopsy specimens were excluded.
Specimens were submitted in either viral transport medium or saline solution for culture, direct immunofluorescence assay, or enzyme immunoassay. Respiratory viruses routinely identified were influenza virus A and B; parainfluenza virus 1, 2, and 3; human respiratory syncytial virus (hRSV); and adenoviruses. Remaining volumes used for this study had been stored at less than –70°C. RT-PCR assays for hMPV were performed in a blind fashion with regards to the patient's medical information.
Case-controlled study. A control group was randomly selected from patients whose samples had tested negative for hMPV and combined with positive patient identifiers on a list which was submitted to two clinical investigators. Seven hMPV-positive patients and four with hMPV-negative results were excluded from the study due to a lack of clinical information. The respiratory samples collected from the control group were retested individually by RT-PCR and confirmed to be hMPV negative. Medical records were examined in a blind fashion, according to a preestablished scheme. The data collection was limited to the information recorded by the treating physicians on medical charts.
The investigators sought the following demographic parameters: age, sex, and postal code. The epidemiologic data included the date of onset of disease, infectious contacts, recent travel history, day care center or school attendance, and hospitalization the week prior to the illness. Underlying medical conditions and medications were reviewed. Investigators also looked for disease manifestations such as fever, rhinorrhea, wheezing, cough, and other signs and symptoms possibly related to the illness.
Results of the following tests were noted: complete blood count, sedimentation rate, chest X-ray, respiratory viral cultures or antigenic detection, nucleic acid detection for Mycoplasma pneumoniae and Bordetella pertussis, and serological studies for specific infectious agents. The need for admission to the intensive care unit, mechanical ventilation, oxygen requirements, and other therapeutic regimens were also recorded.
Nucleic acid extractions. In the first round of screening, all samples except BAL specimens were analyzed in pools of eight samples, prepared by mixing equal volumes of each specimen in the order they were originally submitted to the laboratory. In preliminary analyses, we had found that the efficiencies of recovering and detecting hMPV from respiratory specimens of various natures spiked with in vitro-cultured viruses were similar for all sample types, with the notable exception of BAL fluid specimens (data not shown). Thus, to avoid possible PCR inhibition caused by BAL samples in pools, all BAL specimens were analyzed individually.
The QIAGEN UltraSens kit (QIAGEN, Mississauga, Ontario, Canada), which allows the extraction of total nucleic acids from up to 1 ml of sample volume, was found to be more efficient than several other commercially available nucleic acid extraction kits for recovering hMPV in spiked respiratory specimens (data not shown). Therefore, this method was selected for extracting nucleic acids from pooled samples. In our hands, the QIAamp UltraSens kit coupled with QIAvac purification module 6S gave a higher nucleic acid yield recovery than the centrifugation-based protocol recommended by the manufacturer. Nucleic acids from individual clinical specimens from positive pools were extracted with the QIAamp viral RNA mini kit coupled with QIAvac module 6S as instructed by the manufacturer.
For BAL samples, total RNA was extracted from 200-μl aliquots using the UltraSpec-3 reagent (Biotecx Laboratories, Inc., TX), according to the manufacturer's instructions.
RT-PCR and DNA sequencing. Initial oligonucleotide primer sets were selected on the basis of sequences published by van den Hoogen et al. (23) and available in GenBank. Large genome segments comprising most of the M and F genes could then be amplified and sequenced from eight distinct hMPV isolates cultured in vitro; isolates characterized correspond to CAN98 and CAN99 groups described by Peret et al. in 2002 (20). Subsequent oligonucleotide primers were designed to detect both genogroups.
RT-PCR was performed in a single step using the One-Step RT-PCR kit from QIAGEN as described by the manufacturer. For detecting hMPV in human specimens, one primer pair targeting the M gene and one targeting the M-F intergenic region were used in parallel: Mf1 (5'-ATGTCTGTACTTCCYAA-3') and MReg2 (5'-GCTTATTGCAGCTTCAACAG-3'); and HC 2f (5'-GCACTGTGTGACTTYATGGA-3') and VIRr (5'-ACTTTCCAAGACATTTTTA-3'). Although the first primer pair was found to be more sensitive with serial dilutions of known positive samples, both sets were equally efficient in detecting hMPV in clinical specimens. Cycling conditions were as follows: 50°C for 30 min; 95°C for 15 min; 40 cycles, with 1 cycle consisting of 94°C for 15 s, 52°C for 60 s, and 72°C for 90 s; followed by a final extension at 72°C for 7 min. Amplicons were detected by agarose gel electrophoresis and staining with ethidium bromide.
The hMPV-positive isolates were further characterized by RT-PCR and DNA sequencing using primers targeting the F and M genes. A 1.9-kb fragment spanning the M-F gene region was generated using Mf1 and Fr1 primers (5'-CCYTCAACTTTGCTTAG-3'). Additional primers used for DNA sequencing were as follows: MFIRf (5'-CAARTAAAAATGTCNTGGAA-3'), MF3 (5'-AGAACAGGTTGGTAYAC-3'), HC 7f (5'-AGTAAGAGAGCT GAAAGARTT-3'), and Fr2 (5'-TGCCGCACAACATTTAG-3').
For DNA sequencing, PCR fragments were purified with QIAquick PCR purification columns from QIAGEN as described by the manufacturer, and both strands were sequenced using the BigDye Terminator Cycle Sequencing V 2.0 Ready Reaction kit (PE Biosystems, Foster City, CA). Labeled DNA fragments were separated and detected using ABI genetic analyzer 3100 (Applied Biosystems).
DNA sequence analyses. Base calling and assemblies were performed using the Phed/Phrap software package (University of Washington). Edited sequences were aligned with Clustal W (EMBL, Heidelberg, Germany). Phylogenetic analyses were carried out by using the Phylogeny Inference Package (PHYLIP, version 3.5c; J. Felsenstein, University of Washington) (12). The distance matrix was generated using the DNAdist program (maximum likelihood), and the tree was generated using the neighbor-joining program. Phylogenetic trees were visualized using Treeview (version 1.6.6).
RESULTS
Detection of hMPV in clinical specimens. In 2001, the number of samples submitted for respiratory virus testing to the SJH virology laboratory displayed the usual seasonal pattern. During the winter months, the majority of samples tested positive either by enzyme immunoassay, direct immunofluorescence, and/or tissue culture for hRSV (Fig. 1). Influenza infection peaked in February. On the other hand, adenovirus- and parainfluenza virus-positive samples were found throughout the year.
In our initial screening for hMPV by RT-PCR, the virus was detected in 32 out of 133 pools (24%), and in a single BAL sample out of 74 tested (1%). Specimens from positive pools were then tested separately. Altogether, 41 specimens collected from 37 patients (19 males and 18 females) were found to be positive for hMPV. At least one hMPV-positive sample was found in each positive pool; two and three positive specimens were found in 12.5% and 6.2% of the positive pools, respectively. The two primer sets used for the RT-PCR screening amplified their respective targets from all positive samples.
Similar to hRSV and influenza, the hMPV respiratory infection appeared seasonal, with 37 of 41 positive specimens (90%) collected from January to March (Fig. 2). During this period, 8% of respiratory specimens tested were positive for hMPV.
Sequencing and alignment of specimens. To determine which hMPV genogroup circulated in Montreal, Canada, in 2001, partial sequencing of the M genes of all positive specimens was carried out. Most specimens (73.2%; n = 30) displayed a higher degree of identity with the genogroup B strain Can98-75 (96.3 to 99.6%) than with the prototype hMPV00-1 strain of genogroup A (83.2 to 87.3%) at the nucleotide level (Fig. 3A). Our sequencing data also suggest that each genogroup may be divided into two subgroups. However, no amino acid differences were observed between the two subgroups.
To investigate further the genetic makeup of the circulating genogroups, sequencing of the F gene was undertaken with samples selected from each subgroup. Three samples more closely related to the prototype hMPV00-1 and five samples related to Can98-75 were sequenced. The nucleotide sequence divergence of the F genes of isolates and reference strains was higher than that of the M gene, but the same subgrouping was observed (Fig. 3B).
Clinical manifestations. Clinical data of patients whose respiratory samples tested positive for hMPV were compared to those of patients who tested negative in a case-controlled study. No differences were noted between the two groups with regard to gender, age group, month of the year at disease occurrence, and home address. However, there was a tendency for more chronic illnesses and infectious contacts in group A (hMPV positive) and for more cases of immunosuppression in group B (hMPV negative) (Table 2).
More patients in group A presented with cough and radiographic images of hyperexpansion, peribronchitis, and atelectasis than those in group B. Serological analyses looking for other infectious agents were requested more often in group B than in group A. The duration of admission for hospitalized children did not differ, nor did the duration of stay in the intensive care unit. The need for oxygen and mechanical ventilation or for other treatment was similar in both groups. The final diagnosis was more frequently worded as bronchiolitis or bronchospasm in group A.
Among the 30 hMPV-positive patients, 6 (20%) were coinfected with another commonly identified respiratory virus, but none were coinfected with influenza viruses (Table 3). On the other hand, in the hMPV-negative control group, 13 out of 31 patients (42%) were infected with respiratory viruses likely to be responsible for their symptoms, including 7 who were shown to be infected with an influenza virus. These results are consistent with hMPV being responsible for a certain proportion of the symptoms observed in group A.
In three separate cases of hMPV infection, the exposure was clearly nosocomial. The first case was a 12-year-old girl with encephalopathy and epilepsy who lived in a rehabilitation center. She developed a phlegmy cough, vomiting, and dyspnea with bronchospasm and was admitted to SJH for pneumonia. The second patient was an 11-year-old male with chronic encephalopathy admitted to SJH in January 2000. Approximately 1 year later, while still hospitalized, he developed a viral illness with fever, respiratory symptoms, hyperventilation, and diarrhea. A diagnosis of upper respiratory tract infection was made. The third patient was a 20-month-old infant suffering from congenital lactic acidosis and severe epilepsy. He had been hospitalized since May 2000; in January 2001, he had a fever. The fever resolved, and no precise diagnosis was established at that time. Respiratory samples from the three patients were identified as positive for hMPV by RT-PCR.
Additionally, hMPV contributed to the death of a 3-year-old child handicapped by thoracic asphyxiating dystrophy (Jeune syndrome) who was receiving steroid and oxygen treatment. The patient had been admitted in January 2001 and died 5 days later from multiorganic and respiratory failure. In a separate case, a 9-month-old infant who had traveled to Mali was taken to the emergency room the day he returned to Canada. Symptoms included fever and respiratory and gastrointestinal symptoms of 24-h duration; he had been coughing for 1 week prior to his admission to SJH. This patient had likely been infected with hMPV abroad, while in Africa or on his way back to Canada.
DISCUSSION
We retrospectively assessed the clinical importance of hMPV infection as a cause of ARTI in a pediatric population in Montreal in the Canadian province of Quebec in 2001. We evaluated the incidence over a full year period by RT-PCR testing on nearly all respiratory samples submitted to the virology laboratory of a major pediatric tertiary care center. Samples for which another respiratory virus had already been found were also included. Thus, our approach allowed us to uncover an unusual seasonal pattern and offered the possibility of identifying coinfections. Clinical symptoms associated with hMPV were then compared with those of other respiratory infections in a case-controlled study.
To detect hMPV, we designed RT-PCR assays to amplify viruses of both genogroups and assess the relative sensitivity and specificity of this test using hMPV spiked respiratory samples of different types. Optimized assays were then applied for screening respiratory samples. The rationale for using a molecular approach for this study was based on the following considerations: initial reports of hMPV replication in cell culture suggested that it is a fastidious virus (23); samples analyzed in this study had been subjected to one or more freeze-thaw cycles, for which the effect on hMPV viability was unknown; no antigenic detection was available, and a molecular technique was needed in any case to formally identify the virus. The present study was not aimed at comparing the sensitivity and specificity of our RT-PCR-based assays with those developed by other groups. Nonetheless, it is worth mentioning that the two primer pairs designed for this study detected variants of the two genogroups and that PCR product sequencing confirmed the identity of virus in all hMPV-positive samples.
hMPV was detected in 41 respiratory specimens (3.6% of samples tested) collected from 37 patients. In cases when more than one clinical specimen from a patient tested positive for hMPV, the sequences derived from the specimens were identical. Phylogenic analyses demonstrated a clear dominance of genogroup B in this sample series. As for most respiratory infections, hMPV was mostly found during winter months (90%, or 37 specimens). Interestingly, the virus was not detected in samples collected in November and December, when hRSV was already active (Fig. 1 and 2). In 2001, the same seasonal hMPV occurrence pattern was observed in the neighboring states of the United States and in other regions of Canada (1, 9). In the present study, hMPV was also detected in samples collected in the spring, summer, and autumn, albeit at a low frequency, indicating that hMPV may circulate throughout the year and cause sporadic infections. This finding carries important implications for laboratory investigation of ARTI cases occurring outside the typical respiratory virus season.
NPA was found to be better than throat swabs for hMPV detection. Many respiratory viruses infect the ciliated epithelium of the posterior nasopharynx, and it is generally accepted that NPA collect a superior number of cells than NP washes, NP swabs, and throat swabs do.
The proportion of BAL samples found to be positive for hMPV appeared low at first sight. However, only 21 BAL specimens (28%) were collected between January and March 2001, when hMPV infections peaked, whereas 38% of other specimen types were collected during this 3-month period. Also, we noted a lower efficiency of our hMPV RT-PCR with BAL samples compared with other sample types when spiked with cultured virus (results not shown), a difference that we attributed either to interference with the high content of human nucleic acids in BAL fluids or to residual glutaraldehyde in endoscope channels and in rinsing water that could have impinged on Taq polymerase efficacy (11). Those technical problems may have contributed to an underestimation of the number of BAL fluid samples found positive for hMPV.
Cough (87%), wheezing (40%), and dyspnea (57%) were the symptoms often associated with hMPV infection. Overall, in agreement with the results of other retrospective or prospective studies comparing symptoms of respiratory infections (3, 4, 7, 9, 10, 15-17, 19, 25, 27), our investigation did not reveal a specific symptomatology which could clinically differentiate hMPV from influenza or hRSV infections. Several investigators have found a distinct association between hMPV and bronchiolitis in young children. Our study did not find such a close association, although the terms bronchiolitis and bronchospasm were used more often in the final diagnosis of hMPV-positive cases (40%) than hMPV-negative cases (16%). Moreover, the detection methods used to identify common respiratory viruses are assumed to be less sensitive than the amplification method applied for the detection of hMPV, and we did not seek the presence of other viruses, such as rhinoviruses, coronaviruses, and parainfluenza virus 4, which were likely involved in a certain number of cases. A recent study conducted in France on children admitted for bronchiolitis found RSV to be the predominant causal agent, followed by rhinoviruses and finally by hMPV (5).
Coinfections with hRSV and human parainfluenza viruses were detected (Fig. 3). Mixed viral infections are thought to occur in 3 to 12% of ARTI (26). Thus, our results were not surprising in the context of the peak of respiratory infections during the winter months in Montreal, Canada. In coinfections, a direct contribution of hMPV to an exacerbation of symptoms was not observed, though the low number of cases found during this study preclude any significant conclusion. It is noteworthy that, despite the fact that both groups of patients were recruited from the same period of the year, only patients from the hMPV-negative control group had influenza virus recovered from their respiratory secretions.
This report demonstrates that pooling respiratory samples for hMPV detection is a simple and efficient way for screening a large volume of specimens. This method can be applied for surveillance. This is of particular importance, since the present study and others (9) have documented hMPV infections acquired nosocomially. Therefore, isolation procedures should be taken when hMPV infection is suspected, especially in pediatric units with severely compromised patients and in intensive care units.
Large prospective studies over longer periods of time will eventually help to determine the full clinical spectrum of manifestations associated with this virus, as well as its role in community-acquired and nosocomially acquired respiratory illnesses. Continued surveillance of hMPV infection is warranted to fully understand and to appreciate the clinical importance of this virus.
ACKNOWLEDGMENTS
We thank Andree Falardeau, Jasmine Chamberland, Lyne Desautels, and Regis Cantin for excellent technical assistance; Pierrette Roy for secretarial support; and Donald Murphy, Rejean Dion, Micheline Fauvel, and Peter Barriga for helpful comments and valuable discussions.
REFERENCES
Bastien, N., D. Ward, P. Van Caeseele, K. Brandt, S. H. Lee, G. McNabb, B. Klisko, E. Chan, and Y. Li. 2003. Human metapneumovirus infection in the Canadian population. J. Clin. Microbiol. 41:4642-4646.
Biacchesi, S., M. H. Skiadopoulos, G. Boivin, C. T. Hanson, B. R. Murphy, P. L. Collins, and U. J. Buchholz. 2003. Genetic diversity between human metapneumovirus subgroups. Virology 315:1-9.
Boivin, G., G. De Serres, S. Cote, R. Gilca, Y. Abed, L. Rochette, M. G. Bergeron, and P. Dery. 2003. Human metapneumovirus infections in hospitalized children. Emerg. Infect. Dis. 9:634-640.
Bosis, S., S. Esposito, H. G. Niesters, P. Crovari, A. D. Osterhaus, and N. Principi. 2005. Impact of human metapneumovirus in childhood: comparison with respiratory syncytial virus and influenza viruses. J. Med. Virol. 75:101-104.
Bouscambert-Duchamp, M., B. Lina, A. Trompette, H. Moret, J. Motte, and L. Andreoletti. 2005. Detection of human metapneumovirus RNA sequences in nasopharyngeal aspirates of young French children with acute bronchiolitis by real-time reverse transcriptase PCR and phylogenetic analysis. J. Clin. Microbiol. 43:1411-1414.
Dollner, H., K. Risnes, A. Radtke, and S. A. Nordbo. 2004. Outbreak of human metapneumovirus infection in Norwegian children. Pediatr. Infect. Dis. J. 23:436-440.
Ebihara, T., R. Endo, N. Ishiguro, T. Nakayama, H. Sawada, and H. Kikuta. 2004. Early reinfection with human metapneumovirus in an infant. J. Clin. Microbiol. 42:5944-5946.
Ebihara, T., R. Endo, H. Kikuta, N. Ishiguro, H. Ishiko, M. Hara, Y. Takahashi, and K. Kobayashi. 2004. Human metapneumovirus infection in Japanese children. J. Clin. Microbiol. 42:126-132.
Esper, F., D. Boucher, C. Weibel, R. A. Martinello, and J. S. Kahn. 2003. Human metapneumovirus infection in the United States: clinical manifestations associated with a newly emerging respiratory infection in children. Pediatrics 111:1407-1410.
Falsey, A. R., D. Erdman, L. J. Anderson, and E. E. Walsh. 2003. Human metapneumovirus infections in young and elderly adults. J. Infect. Dis. 187:785-790.
Farina, A., M. H. Fievet, F. Plassart, M. C. Menet, and A. Thuillier. 1999. Residual glutaraldehyde levels in fiberoptic endoscopes: measurement and implications for patient toxicity. J. Hosp. Infect. 43:293-297.
Felsenstein, J. 1997. An alternating least squares approach to inferring phylogenies from pairwise distances. Syst. Biol. 46:101-111.
Freymouth, F., A. Vabret, L. Legrand, N. Eterradossi, F. Lafay-Delaire, J. Brouard, and B. Guillois. 2003. Presence of the new human metapneumovirus in French children with bronchiolitis. Pediatr. Infect. Dis. J. 22:92-94.
Galiano, M., C. Videla, S. S. Puch, A. Martinez, M. Echavarria, and G. Carballal. 2004. Evidence of human metapneumovirus in children in Argentina. J. Med. Virol. 72:299-303.
IJpma, F. F., D. Beekhuis, M. F. Cotton, C. H. Pieper, J. L. Kimpen, B. G. van den Hoogen, G. J. van Doornum, and D. M. Osterhaus. 2004. Human metapneumovirus infection in hospital referred South African children. J. Med. Virol. 73:486-493.
Konig, B., W. Konig, R. Arnold, H. Werchau, G. Ihorst, and J. Forster. 2004. Prospective study of human metapneumovirus infection in children less than 3 years of age. J. Clin. Microbiol. 42:4632-4635.
Mullins, J. A., D. D. Erdman, G. A. Weinberg, K. Edwards, C. B. Hall, F. J. Walker, M. Iwane, and L. J. Anderson. 2004. Human metapneumovirus infection among children hospitalized with acute respiratory illness. Emerg. Infect. Dis. 10:700-705.
Nissen, M. D., D. J. Siebert, I. M. Mackay, T. P. Sloots, and S. J. Withers. 2002. Evidence of human metapneumovirus in Australian children. Med. J. Aust. 176:188.
Peiris, J. S., W. H. Tang, K. H. Chan, P. L. Khong, Y. Guan, Y. L. Lau, and S. S. Chiu. 2003. Children with respiratory disease associated with metapneumovirus in Hong Kong. Emerg. Infect. Dis. 9:628-633.
Peret, T. C., G. Boivin, Y. Li, M. Couillard, C. Humphrey, A. D. Osterhaus, D. D. Erdman, and L. J. Anderson. 2002. Characterization of human metapneumoviruses isolated from patients in North America. J. Infect. Dis. 185:1660-1663.
Stockton, J., I. Stephenson, D. Fleming, and M. Zambon. 2002. Human metapneumovirus as a cause of community-acquired respiratory illness. Emerg. Infect. Dis. 8:897-901.
van den Hoogen, B. G., T. M. Bestebroer, A. D. Osterhaus, and R. A. Fouchier. 2002. Analysis of the genomic sequence of a human metapneumovirus. Virology 295:119-132.
van den Hoogen, B. G., J. C. de Jong, J. Groen, T. Kuiken, R. de Groot, R. A. Fouchier, and A. D. Osterhaus. 2001. A newly discovered human pneumovirus isolated from young children with respiratory tract disease. Nat. Med. 7:719-724.
van den Hoogen, B. G., D. M. Osterhaus, and R. A. Fouchier. 2004. Clinical impact and diagnosis of human metapneumovirus infection. Pediatr. Infect. Dis. J. 23:S25-S32.
Viazov, S., F. Ratjen, R. Scheidhauer, M. Fiedler, and M. Roggendorf. 2003. High prevalence of human metapneumovirus infection in young children and genetic heterogeneity of the viral isolates. J. Clin. Microbiol. 41:3043-3045.
Waner, J. L. 1994. Mixed viral infections: detection and management. Clin. Microbiol. Rev. 7:143-151.
Williams, J. V., P. A. Harris, S. J. Tollefson, L. L. Halburnt-Rush, J. M. Pingsterhaus, K. M. Edwards, P. F. Wright, and J. E. Crowe, Jr. 2004. Human metapneumovirus and lower respiratory tract disease in otherwise healthy infants and children. N. Engl. J. Med. 350:443-450.(Frederic Chano, Celine Ro)